International Workshop on Cross-Language Speech

s
International Workshop on
Cross-Language Speech Perception
3 0 - 3 1 Ja nu ar y 2 0 1 4
U ni v e r si t y o f Mi n ho
B r ag a - P o r t u ga l
International Workshop on Cross-Language Speech Perception
Department of English and North -American Studies (DEINA) | Centre for Humanistic Studies (CEHUM)
Institute of Arts and Human Sciences (ILCH)
GENERAL PROGRAM
January 30, 2014
14:00 Registration
14:30 Opening Ceremony
14:45 Plenary Session – Denise C. Kluge: The role of visual cues in the perception of non-native
sounds
15:45 Session: Ongoing Research (1)
17:05 Coffee break
17:30 Session: L2 Pronunciation Teaching and Miscellanea
19:30 Conference Dinner (not included in the registration fee)
January 31, 2014
09:30 Plenary Session – Ocke-Schwen Bohn: Cross-language speech perception: What we know,
what we'd like to know
10:30 Session: Non-native Speech (1)
12:00 Lunch
13:30 Session: Ongoing Research (2)
14:50 Session: Non-native Speech (2)
16:20 Coffee break
16:45 Keynote Speaker: Márcia C. Zimmer: Some psycholinguistic considerations on methodological
issues related to speech perception tests
17:45 Discussion and meeting about possible partnerships
18:15 Closing Ceremony
PROGRAM
Thursday, January 30
Session
Plenary Session
Ongoing
Research (1)
L2
Pronunciation
Teaching and
Miscellanea
AUDITORIUM (ILCH)
Talks
14:45
The role of visual cues in the perception of non-native sounds
Denise Cristina Kluge
15:45
16:05
16:25
16:45
Crosscultural and
Speaker identification
Listener: a
Perception and
linguistic effects on
in disguised voice and pronunciation training
production of
perception of
imitation by Brazilian
system for BrazilianEuropean Portuguese
European Portuguese
Portuguese listeners
accented English
sounds by Chinese
emotional speech
learners
Denise Kluge, Maria
Gustavo A.
Ana Nunes
Lúcia de Gomes
Mendonça, Sara
Diana Oliveira
Castro
Candeias, Aldebaro
Klautau, Sandra M.
Aluísio
17:05-17:30 - Coffee break
17:30
18:00
18:30
Verbotonal method in
Children’s auditory perceptual
Speech technology products
phonetic correction of palatal performance in the identification derived from the dialog between
lateral approximant /ʎ/
of phonic contrasts
linguists and engineers
Arnalda Dobric
Larissa Cristina Berti
Andréia Schurt Rauber
Friday, January 31
Session
Plenary
Session
Nonnative
Speech
Perception (1)
Ongoing
Research (2)
AUDITORIUM (ILCH)
Talks
9:30
Cross-language speech perception: What we know, what we'd like to know
Ocke-Schwen Bohn
10:30
11:00
11:30
Language-specific
Loanword adaptation is crossInhibitory control and L3
processing of speech and
language speech perception
phonological processing in
non-speech
Catalan/Spanish bilingual
Silke Hamann, Paul Boersma
learners
of
English
Anita Wagner,
Paul Iverson
Joan C. Mora
Isabelle Darcy
12:00-13:30 - Lunch
ROOM 1005 (ILCH)
13:30
13:50
14:10
14:30
Perception and
Cross-language
Assimilation of initial
Moving away from
production of L2
categorization of
English consonants to
repetitions and
English consonants
English vowels by
Danish
discriminations
and vowels: A cross
European Portuguese
training study
perceivers
Camilla Søballe
Yolanda Joy Calvo
in progress
Horslund, Anne
Benzies
Anabela Rato
Aarhøj Andersen,
Angelica Carlet
Ocke-Schwen Bohn
International Workshop on Cross-Language Speech Perception – University of Minho – January 30-31, 2014
-2-
Non-native
Speech
Perception (2)
14:50
The role of language
proficiency in the perception of
L2 voicing contrasts
Grzegorz Aperliński,
Aleksandra Leończyk
Plenary
Session
15:20
The perception of Western
Andalusian Spanish
aspirated stops by General
American English listeners
15:50
Perceptual assimilation of English
vowels and diphthongs by Chilean
Spanish speakers
Juli Cebrian
Maria Del Saz
16:20-16:45
Coffee break
16:45
Some psycholinguistic considerations on methodological issues related to speech perception tests
Márcia Cristina Zimmer
17:45-18:15
Discussion and meeting about possible partnerships
International Workshop on Cross-Language Speech Perception – University of Minho – January 30-31, 2014
-3-
ABSTRACTS
Thursday, January 30
KEYNOTE SPEAKER
The role of visual cues in the perception of non-native sounds
Denise Cristina Kluge (Federal University of Paraná, Brazil)
[email protected]
Considering that human speech is a multimodal function and can be apprehended by visual and auditory means
(Rosenblum, 2005), this talk will present and discuss the role of visual cues in speech perception considering some of
the main studies in the area (e.g., McGurk & McDonald 1976; Grant & Seitz, 1998). This presentation will also review
and discuss some studies that investigate the effect of visual cues in the identification and/or discrimination of nonnative
contrasts (Hazan, Sennema, Iba & Faulkner, 2005; Hazan, Sennema, Faulkner, Ortega-Llebaria, Iba & Chung 2006;
Kluge, Reis, Nobre-Oliveira & Bettoni-Techio, 2009; Kluge, 2009), and will also address some methodological issues
concerning perception tests that make use of visual input.
SESSION: ONGOING RESEARCH (1)
Crosscultural and linguistic effects on perception of European Portuguese emotional speech
Ana Nunes (University of Macau, China)
[email protected]
Research has shown that emotions are not equally recognized. Perceptual tests demonstrated, for example,
identification problems between the pairs neutral/sadness and joy/anger. Joy, sadness and anger can be considered
basic emotions, as they are easier to identify perceptually across different languages and cultures. Other emotions can
be part of the specificities of a Language or of an individual speaker. In order to have a more detailed picture of crosslinguistic effects on emotional speech perception, studies of additional languages are needed. As a contribution to the
area, a perception test was performed using European Portuguese emotional speech. The influence of language and
previous knowledge of cultural aspects was investigated. A new corpus for European Portuguese was recorded having
an actor as informant. In each of two sessions, five repetitions of two semantically emotion neutral sentences (a long and
a short sentence) were recorded using five emotions: joy, sadness, despair, fear, anger, and neutral. Three different
groups participated in the perception (identification) test: two Chinese groups (one formed by Mandarin native speakers
and the other by Cantonese speakers), both were learning Portuguese for three years; a Portuguese group, all
Portuguese native speakers. Scores were compared across language, groups and emotions. Effects of linguistic factors,
such as sentence length, were also analyzed. Main results show that native Portuguese speakers were better at
identifying all emotions, especially sadness. All, except fear, showed statistically significant differences between the
groups; for all the groups anger is the most identifiable emotion; longer sentences increase identification scores in the
three groups.
Audibert, Nicolas; Aubergé, Véronique; Rilliard, Albert, “The Prosodic Dimensions of Emotion in Speech: the Relative weights of Parameters”,
InterSpeech, 2005.
Campbell, Nick; Mokhtari Parham, Voice Quality: 4th Prosodic Dimension, 15th ICPhS, 2003.
Cahn, Janet E., Generating Expression in Synthesized Speech. Master’s Thesis, Massachusetts Institute of Technology, 1989.
Cowie, E. Douglas; Cowie, R. and Schroeder, M., “The description of naturally occurring emotional speech”, 15th ICPhS, 2003.
Johnstone, T. & Scherer, K. R. (1999). “The effects of emotions on voice quality”. Proceedings of the XIVth International Congress of Phonetic
Sciences.
Nunes, Ana Margarida Belém, Voice Quality in European Portuguese, PhD Thesis, University of Aveiro, Portugal, 2009.
Nunes, Ana Margarida Belém, Coimbra, Rosa Lídia, Teixeira, António, Voice Quality of European Portuguese Emotional Speech, Artificial
Intelligence (LNAI) volume, 6001, 2010, 145-151, Springer 2010.
Sawamura, Kanae; Dang, Jianwu; Akagi, Masato; Erickson, Donna; Li, Aijun; Sakuraba, Kyoko; Nimematsu, Nobuaki; Hirose, Keikichi, Common
Factors in Emotion Perception Among Different Cultures, ICPHS XVI, 2007.
Susca, Michael & Healey, E. Charles (2002). “Listener perceptions along a fluency–disfluency continuum: A phenomenological analysis”. Journal of
Fluency Disorders, 27, 135-161.
Rodrigues, Américo, “As emoções na fala [The Emotions in Speech]”, Tese de Mestrado, Universidade de Aveiro, 2007.
Sawamura, Kanae; Dang, Jianwu; Akagi, Masato; Erickson, Donna; Li, Aijun; Sakuraba, Kyoko; Nimematsu, Nobuaki; Hirose, Keikichi, Common
Factors in Emotion Perception Among Different Cultures, ICPHS XVI, 2007.
International Workshop on Cross-Language Speech Perception – University of Minho – January 30-31, 2014
-4-
Speaker identification in disguised voice and imitation by Brazilian Portuguese listeners
Denise Cristina Kluge (Federal University of Paraná, Brazil)
Maria Lúcia de Gomes Castro (Technical Federal University of Paraná, Brazil)
[email protected], [email protected]
This study will present preliminary results in the perception of disguised voice and imitation and it is part of a research
project that aims at describing four acoustic correlates of vowel identity in two different voice qualities: normal and
disguised. The present study aims at investigating whether a group of ten Brazilian Portuguese listeners can
identify/discriminate speakers in normal and disguised voice, as well as in imitation. In order to do so, four discrimination
tests and one identification test were designed using the free software TP (Rauber, Rato, Kluge, Santos, 2012). The
disguised voice’s stimuli were recorded by four men and four women reading a small text that simulated a telephone call
in a kidnapping situation. Speakers were asked to read the text first two times at their normal voice, and then two times
at a freely chosen form of disguise. The stimuli for the tests regarding imitation were recorded by a professional imitator.
Preliminary results from a statistical analysis showed that it was more difficult for the ten listeners to identify than to
discriminate the original voice from the imitation. As regards disguised voice, the results showed that it was more difficult
for the listeners to discriminate the speaker when lip protrusion was the strategy used to disguise the voice.
Rauber, A. S.; Rato, A.; Kluge, D. C., Santos, G. R. dos (2012). TP – Testes de Percepção/Tarefas de Treinamento Perceptual [Perception
Tests/Perceptual Training Tasks], computer program. Available at http://www.worken.com.br/tp_regfree.php
Listener: a pronunciation training system for Brazilian-accented English
Gustavo A. Mendonça1, Sara Candeias2, Aldebaro Klautau3, Sandra M. Aluísio1
1University of São Paulo (ICMC/NILC/USP, Brazil)
2University of Coimbra (IT/UC, Portugal)
3Federal University of Pará (LaPS/UFPA, Brazil)
[email protected], [email protected], [email protected], [email protected]
Recent surveys have shown that Brazil is among the countries with the lowest knowledge of the English language. In
Education First's English Proficiency Index (2012), for instance, Brazil ranked 46th out of 54 countries. In GlobalEnglish
(2013), Brazil achieved a similar position, being ranked as 71st of 77 positions. We aim at improving the proficiency on
the English language among Brazilians, by building an automatic-speech-recognition-based Pronunciation Training
System for Brazilian-accented English. This Pronunciation System, called Listener, will be able to provide online
feedback regarding the pronunciation of the user. Nine types of mispronunciations were selected to be dealt by the
system, according to the study by Zimmer (2004) and those reported in Zimmer et al. (2009), which describe the
pronunciation of English as a foreign language by undergraduate Brazilian students, classified in four levels of
proficiency. We intend to present the Listener’s prototype. It was implemented and tested, using the Hidden Markov
Model Toolkit (HTK) (Young & Young, 1994), and applies Hidden Markov Models (HMM) to represent the acoustic
features of speech. The HMMs are continuous-density Gaussian mixture tied-state triphones with clustering performed
using phonetic decision trees. An excerpt (~2 hours) of the Corpus Oral de Brasileiros Aprendizes de Inglês (COBAI)
was used to build the acoustic model. The system was tested in an isolated words task, by using forced alignment via
Viterbi's algorithm. Pronunciation variants were added to the pronunciation model through symbolic rules. Some
preliminary results related to consonant cluster simplification (“study” > [is.tʌdɪ]) and schwa paragoge (“book” > [bʊ.kɪ])
are analyzed and discussed. Results show that the ongoing method is promising and can allow a wide-ranging
Computer-Assisted Pronunciation Training (CAPT) System.
EducationFirst. (2012). Retrieved Sept. 2nd, 2013, from English Proficiency Index 2012: http://www.ef.com.br/epi/downloads
GlobalEnglish (2012) The Business English Index 2012 Report: Analyzing the Trends of Global Readiness for Effective 21st-Century
Communication. Brisbane: GlobalEnglish.
Young, S., Young, Sj (1994). The HTK Hidden Markov Model Toolkit: Design and Philosophy. In Entropic Cambridge Research Laboratory Ltd, v. 2,
2-44.
Zimmer, M. (2004) A Transferência do Conhecimento Fonético-Fonológico do Português Brasileiro (L1) para o Inglês (L2) na Recodificação
Leitora: Uma Abordagem Conexionista. Dissertação de Doutorado. Porto Alegre: Pontifícia Universidade Católica do Rio Grande do Sul
(Portuguese).
Zimmer, M., Silveira, R., & Alves, U. (2009). Pronunciation Instruction for Brazilians: Bringing Theory and Practice Together. Newcastle: Cambridge
Scholars.
International Workshop on Cross-Language Speech Perception – University of Minho – January 30-31, 2014
-5-
Perception and production of European Portuguese sounds by Chinese learners
Diana Oliveira
[email protected]
Accurate pronunciation is a central research topic in the field of second language (L2) speech acquisition. It is widely
accepted that foreign-accented production [caused by (supra)segmental deviations] derives mainly from perceptual
inaccuracy rather than from articulatory difficulties [1; 2]. From childhood to adulthood, exposure to first language (L1)
speech sounds reduces perceptual sensitivity to phonetic/acoustic categories with no functional status in the L1. This
attunement to L1 sounds leads to a biased perception which, in turn, origins an accented production [3]. Therefore,
research has focused on the perceptual phenomena underlying acquisition of L2 speech. Several studies, targeting
different L1-L2 pairings, have investigated the efficiency of laboratory perceptual training. It has been shown that: i)
short-term training is effective in improving listeners’ perception skills [4; 5]; ii) perceptual learning is transferrable to new
tokens and new talkers and long-term retention is possible [5; 6]; iii) enhancement in perception tends to correlate
positively with performance in production [6; 7]. However, the success of perceptual training is known to vary according
to the specific L1-L2 pair under study, depending on the phonological/phonetic and phonotactic (dis)similarities between
the two languages [3]. This study aims at testing the efficiency of perceptual training in the improvement of perception
and production of European Portuguese (EP) consonant sounds by Mandarin speakers, who are currently an important
group of EP learners [8; 9]. To our knowledge, no experimental investigation to date has studied the combination of
Mandarin as L1 and EP as L2. Thus, it is important to understand: i) which EP sounds are incorrectly produced by
Mandarin speakers; ii) whether or not and to what extent previous results of successful training procedures can be
accomplished for this specific L1-L2 pairing; iii) how resulting knowledge can be made useful for designing adequate
EPL2 teaching materials.
1. Flege, J. (1995). Second Language Speech Learning: Theory, Findings and Problems. In Strange, W. (Ed), Speech Perception and Linguistic
Experience: Issues in Cross Language Research (pp. 233-277). Timonium, MD: New York Press.
2. Best, C. & Tyler, M. (2007). Nonnative and Second Language Speech Perception: Commonalities and Complementarities. In Bohn, O. & Munro,
M. (Eds), Language Experience in Second Language Speech Learning – In Honor of James Emil Flege (pp. 13-34).
Amsterdam/Philadelphia: John Benjamins Publishing Company.
3. Strange, W. (1995). Cross language studies of speech perception - a historical review. In Strange, W. (Ed), Speech Perception and Linguistic
Experience: Issues in Cross Language Research (pp. 3-45). Timonium, MD: New York Press.
4. Rochet, B. (1995). Perception and Production of Second Language Speech Sounds by Adults. In Strange, W. (Ed), Speech Perception and
Linguistic Experience: Issues in Cross Language Research (pp. 379-410). Timonium, MD: New York Press.
5. Wang, X. (2008). Perceptual Training for Learning English Vowels – Perception, Production and Long-Term Retention. Saarbrücken, Germany:
Verlag Dr. Müller.
6. Yamada, R., Tohkura, Y., Bradlow, A. & Pisoni, D. (1996). Does Training in Speech Perception Modify Speech Production?. Proceedings of the
Fourth International Conference on Spoken Language Processing, v. 2 (pp. 606-609). Philadelphia: Institute of Electrical and Electronics
Engineers.
7. Alliaga-García, C. & Mora, J. (2009). Assessing the effects of phonetic training on L2 sound perception and production. In Watkins, M., Rauber,
A. & Baptista, B. (Eds), Recent Research in Second Language Phonetics/Phonology – Perception and Production (pp. 2-31). Newcastle
upon Tyne: Cambridge Scholars Publishing.
8. Reto, L. (Coord) (2012). Potencial Económico da Língua Portuguesa. Alfragide: Texto Editores.
9. Belchior, M. (2011). Chineses aprendem mais português. Revista Macau (23). Retrieved May 24, 2013, from
http://www.revistamacau.com/2011/06/05/chineses-aprendem-mais-portugues/
SESSION: L2 PRONUNCIATION TEACHING AND MISCELLANEA
Verbotonal method in phonetic correction of palatal lateral approximant /ʎ/
Arnalda Dobric (Faculty of Humanities and Social Sciences, University of Zagreb)
[email protected]
Although often neglected in foreign language teaching, pronunciation holds a very important role in language acquisition
or learning. On the phonological level adult learners of foreign languages tend to have problems with the speech sounds
that are close to some sounds in their native language but learners do not recognize them as different and replace them
with the similar speech sounds from their native language. These errors are almost systematic if learners belong to the
same L1 and learn the same L2. In the verbotonal method it is referred to as the system of errors. Another set of
problems arises with the sounds that do not exist in the learners’ native language. In this paper, effects of phonetic
training based on several elements (sound context with respect to optimal frequency band and coarticulation, position of
International Workshop on Cross-Language Speech Perception – University of Minho – January 30-31, 2014
-6-
the target sound within the word or sentence, intonation, tenseness of articulation as well as movement and body
posture accompanying speech production) according to the verbotonal method are shown in the example of the palatal
lateral approximant /ʎ/, which is one of the hardest sounds to pronounce for foreign learners of Croatian as L2. Learners
of different native languages (Spanish, Czech, English, German and Hungarian) were recorded before and after
sessions of phonetic training. The results were assessed using a perceptual test. Furthermore, acoustical analyses were
performed in Praat. The results showed higher F2 after correction sessions. Since body posture accompanying speech
production seems to be one of the crucial elements for correction of this specific error, the importance of body movement
and posture in speech is discussed in more detail. Overall changes in pronunciation after phonetic training are also
discussed.
Children’s auditory perceptual performance in the identification of phonic contrasts
Larissa Cristina Berti (Paulista State University, Brazil)
[email protected]
The aim of this study was to assess the auditory perceptual performance of children in a task of identification of phonic
contrasts in Brazilian Portuguese (henceforward BP), and to classify which phonic classes provide higher or lower
degrees of difficulty. A forced-choice minimal-pair identification task involving the 19 phonemes in BP was conducted
with 66 children, 4-5 years old, using PERCEVAL software. The stimuli used in the identification task consisted of a
typical adult’s recordings of the familiar disyllabic words (minimum pairs) contrasting the 19 phonemes in stressed
position. The acoustic stimulus was presented to the children using headphones and they needed to choose which
stimulus-correspondent picture was shown on the computer screen. Both presentation time and reaction time of the
stimulus were measured by PERCEVAL software. Our results showed that the perceptual accuracy was dependent on
phonic classes according to the following decreasing order: vowels (d’ prime = 1,73) > sonorants (d’ prime = 1,56) >
occlusives (d’ prime = 1,43) > fricatives (d’ prime = 0,93). The reaction time of the incorrect response was significantly
higher than the reaction time of the correct response in all the phonic classes. Within each class, there are contrasts that
present a greater difficulty in the identification task. Altogether, the perceptual difficulty varies due to phonetic distance.
The auditory perceptual mastering of the phonic contrasts in BP occurs gradually and, within each class, there seems to
be a hierarchy in the perceptual mastering. The parallel between perceptual acquisition orders and production
acquisition orders was not always the same.
Speech technology products derived from the dialog between linguists and engineers
Andréia Schurt Rauber (Appen Butler Hill)
[email protected]
The partnership between linguists, engineers and programmers has become crucial in order to design speech products
that have increasingly become part of our daily lives. Voice commands have become easily available in cars,
smartphones and smart TVs, for example, but how does the machine recognize human speech, process requests and
react to users? In this talk I will present linguistic tasks that are essential for the creation of speech recognizers and
synthesizers, and will also stress that the knowledge of first and foreign language phonology, and (acoustic) phonetics is
necessary for engineers and programmers to develop the voice-operated apps that surround us. Three questions
motivate this talk: (1) are phonologists/phoneticians aware of their importance for the creation of speech products? (2)
Are linguists open to interact with other fields and also go beyond academic projects? (3) What is the role of speech
perception in the design of synthesizers that use the phonological inventory of a given language to synthesize words in
other languages?
International Workshop on Cross-Language Speech Perception – University of Minho – January 30-31, 2014
-7-
Friday, January 31 - morning
KEYNOTE SPEAKER
Cross-language speech perception: What we know, what we'd like to know
Ocke-Schwen Bohn (Aarhus University, Denmark)
[email protected]
The perception of nonnative speech sounds is guided by two kinds of biases: Those which can be attributed to the native
sound system, and those which reveal universal preferences for particular acoustic dimensions or types of sounds. This
presentation will briefly review the success with which the most widely used models of second language (L2) speech,
Flege’s SLM (1995) and Best’s PAM (1995) and PAM-L2 (Best & Tyler 2007) predict the application of L1 biases to L2
perception, and it will then attempt to address open questions regarding L1 biases, such as how to assess crosslanguage similarity of speech sound categories, and questions surrounding the role of higher-order phonological biases
in cross-language speech perception. The presentation will also review two universal biases in cross-language speech
perception which, respectively, provide the basis for Bohn’s (1995) desensitization hypothesis and for Polka and Bohn’s
(2003, 2011) Natural Reference Vowel framework. These well-documented biases inspire questions regarding less-well
studied phenomena in consonant and vowel perception which might reveal universal preferences.
Best, Catherine T. 1995. A direct realist view of cross-language speech perception. In: Strange, W., ed., Speech perception and linguistic
experience: Issues in cross-language research. Timonium, MD: York Press. 171-204.
Best, Catherine T. & Tyler, Michael D. 2007. Nonnative and second language speech perception. In: Bohn, Ocke-Schwen & Munro, Murray J., eds.
2007. Language experience in second language speech learning: In honor of James Emil Flege. Amsterdam & Philadelphia: John Benjamins,
13-34.
Bohn, Ocke-Schwen. 1995. Cross-language speech perception in adults: First language transfer doesn’t tell it all. In: Strange, W., ed., Speech
perception and linguistic experience: Issues in cross-language research. Timonium, MD: York Press, 279-304.
Flege, J. E. 1995. Second language speech learning: Theory, findings, and problems. In: Strange, W., ed., Speech perception and linguistic
experience: Issues in cross-language research. Timonium, MD: York Press. 233-277
Polka, Linda & Bohn, Ocke-Schwen. 2003. Asymmetries in vowel perception. Speech Communication 41, 221-231.
Polka, Linda & Bohn, Ocke-Schwen. 2011. Natural Referent Vowel (NRV) framework: An emerging view of early phonetic development. Journal of
Phonetics 39, 467-478.
SESSION: NONNATIVE SPEECH PERCEPTION (1)
Language-specific processing of speech and non-speech
Anita Wagner (University Medical Center Groningen, The Netherlands)
Paul Iverson (University College London, UK)
[email protected], [email protected]
Adult listeners perceive speech sounds in language-specific ways. It is an open question what levels of processing are
adapted to the mother tongue, but it is generally assumed that language experience does not alter auditory processing.
We investigated language-specific sensitivities to acoustic differences between English /r/ and /l/ in Japanese listeners,
and studied the processing of these sounds at preattentive auditory stages and at levels of phonetic categorization.
When listening to speech, Japanese listeners show language-specific sensitivities to the acoustic dimensions that
differentiate English /r/ from /l/. When presented with the same acoustic variation in non-speech contexts, however,
Japanese listeners show a pattern of discrimination not different from English. The present study used stimuli that
parametrically varied this speech/nonspeech difference, to see how far a stimulus must be from speech in order for
cross-language differences to disappear. The experimental stimuli were a series of continua that mimicked the acoustic
differences between /r/ and /l/. The different continua disrupted, to varying degrees, the naturalness and perceptual
coherence of the stimuli. Behavioral results demonstrated strong cross-language differences for the stimuli that sounded
most like speech, no significant cross-language differences for the stimuli that were least like speech, and moderate
cross-language differences for a middle range of stimuli that were not categorized accurately as /r/ and /l/ and did not
sound like speech. The cross-language differences for the latter stimuli were not caused by phonological categorization,
but they may be somewhat speech specific. An MEG experiment recorded listeners’ mismatch responses (MMF) to four
stimulus series, which showed different degrees of language-specific processing. These results show cross-language
differences in the processing of all four transforms. Stimuli that defied phonetic categorization showed language-
International Workshop on Cross-Language Speech Perception – University of Minho – January 30-31, 2014
-8-
dependent processing at a low pre-attentive level. This suggests that language-specialization affects the processing of
acoustic variation for both speech and non-speech.
Inhibitory control and L3 phonological processing in Catalan/Spanish bilingual learners of English
Joan C. Mora (Universitat de Barcelona, Spain)
Isabelle Darcy (Indiana University, USA)
[email protected], [email protected]
Recent research suggests that L2 learners with better inhibitory control may be more successful at keeping their L1 and
L2 phonological systems apart, thus avoiding cross-language phonological interference in speech perception and
production (Darcy, Mora, Daidone, in press; Lev-Ari, Peperkamp, 2013). The present study investigates whether
Catalan/Spanish bilingual learners of English with better inhibitory skill develop more accurate representations for L3English sounds. Inhibition was assessed through a retrieval-induced inhibition task (Lev-Ari, Peperkamp, 2013; Veling,
van Knippenberg, 2004). L3-English phonology was assessed through a speeded ABX categorization task (perception)
and a delayed sentence repetition task (production), while partialling out proficiency through an L3 vocabulary size
measure. Participants were early Catalan/Spanish bilinguals, who were also late learners of English (their L3). The
participants used both Catalan and Spanish daily, but crucially differed in the amount of use of their less dominant
language (either Catalan or Spanish). They were assigned to two groups according to the percentage of use of their less
dominant language: “unbalanced” (<30%) vs. “balanced” (>30%). We predicted that the unbalanced group, unlike the
balanced group, would have developed a stronger inhibitory control due to their extensive experience in inhibiting their
more dominant language when using their less dominant one. We further hypothesized that their enhanced inhibitory
control would be related to more efficient L3 (English) phonological processing. Results showed that the “unbalanced”
group indeed had a significantly stronger inhibitory control than the “balanced” group, in the way we had predicted.
However, inhibitory control scores were unrelated to L3 phonological scores. We tentatively concluded that the effects of
individual differences in inhibitory control in these early bilinguals might have been “washed out” due to the daily practice
they receive in inhibiting one language over the other, which calls for further research on the relationship between
inhibitory control and L2 and L3 phonological processing.
Darcy, I., Mora, J. C. & Daidone, D. (in press). Attention control and inhibition influence phonological development in a second language. In
Proceedings of the International Symposium on the Acquisition of Second Language Speech. Concordia Working Papers in Applied
Linguistics, 5.
Lev-Ari, S., & Peperkamp, S. (2013). Low inhibitory skill leads to non-native perception and production in bilinguals’ native language. Journal of
Phonetics,41, 320-331.
Veling, H. & van Knippenberg, A. (2004). Remembering can cause inhibition: Retrieval-induced inhibition as cue independent process. Journal of
Experimental Psychology: Learning, Memory and Cognition 30, 2, 315-318).
Loanword adaptation is cross-language speech perception:
case studies from Malayalam, Thai and Japanese
Silke Hamann, Paul Boersma (University of Amsterdam, The Netherlands)
[email protected], [email protected]
Studies on loanword adaptation (such as Davidson & Noyer, 1996; Broselow, 2003; Yip, 2006) usually employ loanwordspecific devices to deal with what Kenstowicz (2005) called “divergent repairs”: loanword adaptations differ from what we
would expect on the basis of the native phonology. In the present account of divergent repairs we follow Boersma and
Hamann (2009) by assuming that loanwords are integrated into a language via native-language perception of nonnative
sounds and forms. This perception process involves both language-specific perceptual cues and structural restrictions,
and can be formalized in a perception grammar (Boersma, 2000, see lower left of the figure) as the interaction of cue
and structural constraints:
International Workshop on Cross-Language Speech Perception – University of Minho – January 30-31, 2014
-9-
In native language phonological production (upper right part of the figure), on the other hand, the same structural
constraints interact with phonological faithfulness constraints. In Malayalam (Mohanan & Mohanan, 2003), for example,
the structural constraint “no voiceless singleton intervocalically”, which is unviolated in the language, is satisfied by
gemination in loanword perception, but by voicing in native phonological production. The English word baker enters the
Malayalam ear as the foreign phonetic form [beikəɹ] and is converted natively to the surface structure /.beːk.kar./
because the cue constraint *[no periodicity]/ voi/ (“phonetic non-periodicity does not correspond to phonological voicing”)
outranks the cue constraint *[small duration]/ long/ (“phonetic shortness does not correspond to a phonological long
segment”). In Malayalam phonological production, the same structural constraint interacts with the faithfulness
constraints IDENT(length) and the lower-ranked DEP(voi), which are responsible for the realization of underlying
intervocalic singleton |makan| as voiced /.ma.gan./. The divergent repairs are thus caused by the different types of
constraints that the structural constraint fights with: faithfulness in production and cues in perception. We will present
further illustrations of divergent repairs from Thai and Japanese, and conclude that loanword adaptation is often just
cross-language speech perception.
Friday, January 31 - afternoon
SESSION: ONGOING RESEARCH (2)
Perception and production of L2 English consonants and vowels: A cross training study
in progress
Angelica Carlet (Universidad Autonoma de Barcelona, Spain)
[email protected]
Foreign language learning in a formal setting is notably characterized by limited target language exposure. Thus, phonetic
training may play a very important role in pronunciation learning, as it can make up for the absence of regular L2
experience. Several studies have shown the efficacy of phonetic training on the ability to correctly perceive and produce L2
consonant sounds (Hardison, 2003; Iverson & Evans, 2007; Pisoni, Lively & Logan, 1994; Strange & Dittmann, 1984,
among others) and L2 vowel sounds (Aliaga-García, 2011; Aliaga-García & Mora, 2009; Ceñoz & Garcia Lecumberri, 1999;
Iverson & Evans, 2007, among others). This study will further investigate the effects of a high variability phonetic training
approach (Logan, Lively & Pisoni, 1991) on the perception and production of non-native sounds, specifically of the English
stop consonant sounds and the English vowel sounds [/æ/-/ʌ/,/ɪ/-/i/] by Spanish/Catalan speakers. Two experimental
groups (N=30) and a control group (N=20) will be exposed to a perceptual cross-training methodology and will be tested on
both their perception and production of L2 sounds by means of a pretest/posttest/retention-test basis. Each experimental
group will be explicitly trained (instruction and feedback) to better identify either vowels or consonants embedded in nonce
words, while being implicitly exposed to the untrained segments (vowels or consonants). This study also aims at
investigating if improvement generalizes to aspects not explored in the training, such as new words, new talkers and to
“implicitly trained” segments. The predictions are that phonetic training will improve L2 performance for the explicitly trained
segments and to a lesser extent to the “cross-trained” segments, due to the lack of focus on specific cues and feedback.
Moreover, generalization is predicted to occur, as well as retention of the knowledge acquired during training.
International Workshop on Cross-Language Speech Perception – University of Minho – January 30-31, 2014
-10-
Cross-language categorization of English vowels by European Portuguese perceivers
Anabela Rato (Universidade do Minho, Portugal)
[email protected]
Cross-language research on phonological learning has revealed that the perception of nonnative vowel contrasts is
highly challenging for adult L2/FL learners mainly due to perceptual biases caused by L1 interference [1]. Studies on
English vowel categorization by perceivers from small-vowel-set romance languages such as Spanish [2, 3, 4], Italian [5,
6], and Portuguese [7, 8, 9] have shown that perceptual patterns in the mapping of English vowels onto L1 phonological
categories are, to a great extent, similar. However, it is important to further investigate each language variety so that
specific perceptual difficulties can be predicted accurately. Therefore, although there are some studies that investigated
the categorization of English vowels by native Brazilian Portuguese (BP) listeners, research regarding the assimilation of
L2 vowels by European Portuguese (EP) perceivers is scarce [10, 11]. In order to assess cross-linguistic perceptual
(dis)similarity between the English and the EP vowel inventories and to further understand reported inaccuracy in the
assimilation of nonnative vowel contrasts, two experiments were carried out with 68 Portuguese learners of English as a
Foreign Language (EFL).The categorization of a set of English vowels was tested with two identification tasks. In
Experiment 1, a group of 34 participants was tested on native and nonnative vowels with an 8AFC cross-language
perceptual assimilation task (PAT), in which listeners identified the target segments in terms of L1 vowel categories and
then rated each segment for goodness-of-fit. In Experiment 2, the other group of 34 students was tested exclusively on
the set of nonnative vowels with a 7AFC ID task, by labelling them according to the target L2 categories. In this paper,
the results of the two experiments are discussed by underlying the L1-L2 (dis)similarities, and a comparison amongst the
results of the aforementioned studies and the findings of these two experiments is made.
1. Strange, W. (2007). Cross-language phonetic similarity of vowels: Theoretical and methodological issues. In O.-S. Bohn, & M. Munro (Eds.),
Language Experience in Second Language Speech Learning – In honor of James Emil Flege (pp. 35-55). Amsterdam/Philadelphia: John
Benjamins Publishing Company.
2. Cebrian, J. (2006). Experience and the Use of Duration in the Categorization of L2 Vowels. Journal of Phonetics, 34, 372-387.
3. Flege, J., Bohn, O.-S. & Jang, S. (1997). Effects of experience on non-native speakers’ production and perception of English vowels. Journal of
Phonetics, 25, 437-470.
4. Flege, J., Munro, M., & Fox, A. (1994). Auditory and categorical effects on cross-language vowel perception. Journal of the Acoustical Society of
America, 95(6), 3623-3641.
5. Flege, J., MacKay, I., & Meador, D. (1999). Italian speakers’ perception and production of English vowels. Journal of the Acoustical Society of
America, 106(5), 2973-4.
6. Flege, J., & MacKay, I. (2004) Perceiving vowels in a second language. Studies in Second Language Acquisition, 26, 1-34.
7. Major, R. (1987). Phonological Similarity, Markedness, and Rate of L2 Acquisition. Studies in Second Language Acquisition, 9(1), 63-82.
8. Nobre-Oliveira, D. (2007). The Effect of Perceptual Training on the Learning of English Vowels by Brazilian Portuguese Speakers (Doctoral
dissertation).
9. Rauber, A. S. (2010). Acoustic characteristics of Brazilian English vowels: perception and production results. Saarbrücken: Lambert Academic
Publishing.
10. Flege, J. (1995). Second Language Speech Learning: Theory, Findings and Problems. In W. Strange (Ed.), Speech Perception and Linguistic
Experience: Issues in Cross Language Research (pp. 233-277). Timonium, MD: York Press.
11. Rato, A., Rauber, A., Soares, L., & Lucas, L. (2013). Challenges in the perception and production of English front vowels by native speakers of
European Portuguese. Manuscript submitted for publication.
Assimilation of initial English consonants to Danish: Identifying areas of expected learning ease and difficulty
for native Danish speaking learners of English
Camilla Søballe Horslund, Anne Aarhøj Andersen,
Ocke-Schwen Bohn (Aarhus University, Denmark)
[email protected], [email protected], [email protected]
Several studies suggest that second language (L2) learners face different challenges depending on how their native
language (L1) maps onto the target language (e.g., Flege et al., 1997), and hence leading models of L2 speech
perception and production take L1-L2 assimilation as their starting point. The Speech Learning Model (SLM) predicts
learning success for the production of individual L2 phones based on their perceived similarity to phones in the learner’s
L1, dividing the L2 phones into new, similar and identical phones vis-á-vis the L1 (Flege 1995). Likewise, the Perceptual
Assimilation Model for L2 learners (PAM-L2) predicts discrimination performance of pairs of L2 sounds based on how the
relationship between the two L2 phones is assimilated to the native phoneme inventory; i.e., whether the two L2 phones
are assimilated to the same L1 category or to different L1 categories as well as the L2 phones’ perceived goodness of fit
as instances of the L1 category (Best & Tyler, 2007). This study investigates how initial English consonants are
International Workshop on Cross-Language Speech Perception – University of Minho – January 30-31, 2014
-11-
assimilated to the phoneme inventory of Danish. Twenty initial English consonants were presented in a /Ca/ frame and
listeners were asked to map them onto one of 15 response alternatives representing initial Danish consonants and
consonant clusters. Listeners also rated the goodness of fit of each assimilation on a 9-point Likert scale. The results of
the study thus present the first step towards predicting learning ease and difficulty of individual English consonants for
L1 Danish learners of English. The talk will present the assimilation results as well as categorise initial English
consonants in terms of SLM and PAM-L2 for L1 Danish learners of English. The talk will also present hypotheses
regarding the acquisition of initial English consonants by L1 Danish speakers based on these categorisations.
Flege, J. E. 1995. Second language speech learning: Theory, findings, and problems. In: Strange, W., ed., Speech perception and linguistic
experience: Issues in cross-language research. Timonium, MD: York Press. 233-277.
Flege, J.E., Bohn, O.-S. and Jang, S. 1997. Effects of Experience on Non-Native Speakers’ Production and Perception of English Vowels”. Journal
of Phonetics. 25, 437-470.
Best, C. T. & Tyler, M. D. 2007. Nonative and second language speech perception. In: Bohn, O.-S. & Munro, M. J., eds. 2007. Language
experience in second language speech learning: In honor of James Emil Flege. Amsterdam: John Benjamins, 13-34.
Moving away from repetitions and discriminations: introducing engaging and motivating activities into EFL
pronunciation classes to help Spanish students improve their perception and production of English vowels
Yolanda Joy Calvo Benzies (Universidad de Compostela)
[email protected]
Spanish learners of English, even those at advanced levels, tend to have serious problems with the oral component and,
more particularly, with pronunciation (Alcaraz & Moody, 1999; Palacios, 2000; Martínez, Usó, &Alcón, 2006). Generally
speaking, these difficulties are due to the many differences between the phonological systems of both languages and to
the low degree of exposure that students from this country have to spoken English outside the classroom. Hence, it
would seem feasible for pronunciation to be emphasized in English as a foreign language (EFL) classes in Spain.
However, even though students consider that pronunciation should receive more attention in their EFL classes (Calvo,
2013a, b), the truth is that this language area continues to receive less attention than other language fields such as
grammar or vocabulary. Moreover, the majority of activities present in EFL textbooks addressed at Spanish learners are
mainly simple drills, repetitions and discrimination exercises (Calvo, 2013c), without engaging the students into real and
authentic oral communication. This paper aims at helping Spanish students improve both their perceptual and productive
levels of pronunciation by: a) describing some of the most modern and authentic materials available for teaching foreign
languages such as restaurant menus, podcasts, maps and songs; and, b) providing ideas and activities to generally
focus on oral skills and, more specifically, on the pronunciation of English vowels, by using the materials mentioned
above. These ideas will address the main difficulties that Spanish learners tend to have with English vowels, such as the
distinctions between short and long vowels, schwa or /^/; authentic materials and topics that students are likely to use in
daily-life situations, such as, food, health and travelling, have been chosen so as to emphasise real oral communication,
rather than the artificial dialogues generally present in EFL materials addressed at Spanish learners.
Alcaraz, Enrique and Moody, Brian (1999). Fonética inglesa para españoles. Alcoy: Marfil.
Calvo Benzies, Yolanda Joy (2013a). "Not Enough Time is Devoted to Pronunciation, I would Like my Textbook to Include other Types of
Pronunciation Activities. The Views and Perspectives of EFL Fourth Year Secondary-school Students on the Teaching and Learning of
Pronunciation. A Survey-based Study" in Mompeán, J.A. and Fouz, Jonás (eds.), Proceedings of the Third International Conference on English
Pronunciation: Issues & Practices (EPIP3). Murcia: Universidad de Murcia, pp. 10-13.
Calvo Benzies, Yolanda Joy (2013b). "Spanish EFL University Students’ Views on the Teaching of Pronunciation: A Survey-Based Study" in Ciarlo,
C and Giannoni, D.S (eds.). Language Studies Working Papers, vol. 5. Reading: University of Reading, pp. 41-49.
Calvo Benzies, Yolanda Joy (2013c). "La enseñanza de la pronunciación del inglés en Galicia: una asignatura pendiente" in Cabedo Nebot, A.;
Aguilar Ruiz, M.J; and López-Navarro Vidal, E. (eds.), Estudios de Lingüística: investigaciones, propuestas y aplicaciones. Valencia:
Universitat de Valencia, pp. 13-20.
Martínez-Flor, A., Usó-Juan, E. and Alcón Soler, E (2006). “Towards acquiring communciative competence through speaking” in Usó-Juan, E. and
Martínez-Flor, A. (eds). Current trends in the development and teaching of the four language skills. Berlin: Mouton de Gruyter, pp, 139-157.
Palacios Martínez, I.M (2000). “Improving learner´s pronunciation of English: some reflections and some practical tips” in Ruiz, José Mª; Sheerin,
Patrick H. and Estebánez, Cayetano (eds.). Estudios de metodología de la lengua inglesa, pp, 17-40.
International Workshop on Cross-Language Speech Perception – University of Minho – January 30-31, 2014
-12-
SESSION: NON-NATIVE SPEECH PERCEPTION (2)
The role of language proficiency in the perception of L2 voicing contrasts
Grzegorz Aperliński (Adam Mickiewicz University, Poznań)
Aleksandra Leończyk (University of Szczecin)
[email protected], [email protected]
English and Polish discriminate between voiced and voiceless pairs of consonants, mostly on the basis of voice onset
time (VOT). English is an aspirating language (Lisker & Abramson, 1964), whereas Polish is a voicing language
(Keating, 1980). Taking this into consideration, one might conclude that English voiced and Polish voiceless obstruents
should sound the same to a naïve Polish listener. A previous study on the perception of voicing revealed that when
English voiceless and voiced obstruents were acoustically manipulated to have the same VOT, Polish listeners correctly
identified English voiced obstruents in 80% of the cases, while English listeners did so nearly 100% of the time
(Aperliński, 2012). While Polish listeners were shown to differ significantly from English listeners in terms of voicing
perception, no difference was found for Polish listeners with different proficiency levels (beginner vs. advanced). The
authors believe that the observed tendency might not have been necessarily due to the lack of differences between the
groups but due to methodological considerations, i.e., the use of self-reported proficiency. The current study aims at
further exploring the L2 perception of voicing with an improved method for testing the participants’ proficiency. Over 60
participants divided into two groups differing significantly in their language proficiency performed an identical task to that
in Aperliński (2012), i.e., forced-choice identification task on 21 sets of English minimal pairs, e.g., pit/bit. The tokens
were acoustically manipulated so that both words in a set had the same short lag VOT. However, the careful control over
their language skills (testing rather than self-reporting) should provide better insights into the role of language
proficiency. If highly proficient Polish listeners display tendencies similar to those of native speakers of English, the
observed results might be interpreted as evidence to the acquisition of L2 phonology.
Aperliński, G. 2012. Is VOT enough? (Paper presented at the VIIth International Conference on Native and Non-native Accents of English, 6-8
December 2012.).
Keating P. A. 1980. A Phonetic Study of a Voicing Contrast in Polish [Unpublished Ph.D. dissertation, Brown University.].
Lisker, L. and Abramson, A. S. 1964. A cross-language study of voicing initial stops: Acoustic measurements. Word 21: 384–422.
The perception of Western Andalusian Spanish aspirated stops by General American English listeners
Maria Del Saz (Universidad de Sevilla, Spain)
[email protected]
In Western Andalusian Spanish (WAS), syllable- and word-final -s is realized as aspiration [h]. This phenomenon is not
limited to southern Spain; it is also present in other areas such as the Canary Islands or the Caribbean and the Pacific
countries. After aspiration, voiceless stops become aspirated [ph, th, kh] (Alvar, 1996; Gerfen, 2002; Torreira, 2007a,
2012). In the case of WAS, these stops are characterized by longer post-aspiration and shorter pre-aspiration than in
other Spanish aspirating dialects (Torreira, 2007b). This study explores the perception of aspirated –s before this set of
stops by 84 General American English (GAE) listeners, learners of Spanish, and aims at pointing the acoustic
characteristics that favor the identification of such underlying –s. Stimuli consisted of nouns and verbs ending in the
morphological marker –s or in vowel, followed by different phonetic contexts, embedded in sentences. These sentences
were presented randomly in an identification task with two forced-choice alternatives. Results of the identification of
aspirated –s before the voiceless stops indicate that listeners identified [th] significantly better than [kh] (Z = -4.696, p <
.001), which in turn was better identified than [ph] (Z = -3.826, p < .001). An acoustic analysis of the target stops in terms
of previous vowels (pre-aspiration) and Voice Onset Time (VOT) (postaspiration) revealed no significant differences in
VOT duration between the three aspirated stops [χ2(2) = -.512, p = .77], while significant differences in the duration of
their preceding vowels appeared [χ2(2) = 6.03, p < .05], which were longer before [th] than before the other two stops.
Conversely, vowels before unaspirated [t] were particularly shorter than before the other two unaspirated stops [χ2(2) =
6.53, p < .05]. From this analysis, it seems that the longer pre-aspiration observed before [th] was a key factor in
identification.
Alvar, M. (1996) Manual de Dialectología Hispánica. El Español de España. Barcelona: Ariel.
Gerfen, C. (2002) Andalusian Codas. Probus, 14,(2), 247-277.
Torreira, F. (2007a) Pre- and post-aspirated stops in Andalusian Spanish. In Pilar Prieto, Joan Mascaro & Maria-Josep Solé (Eds.), Prosodic and
Segmental Issues in Romance, (pp. 67–82). Amsterdam: John Benjamins.
International Workshop on Cross-Language Speech Perception – University of Minho – January 30-31, 2014
-13-
Torreira, F. (2007b) Coarticulation between aspirated-s and voiceless stops in Spanish: An interdialectal comparison. In N. Sagarra & A. J. Toribio
(Eds.), Selected Proceedings of the 9th Hispanic Linguistics Symposium, (pp. 113–120). Somerville, MA: Cascadilla Press.
Torreira, F. (2012) Investigating the nature of aspirated stops in Western Andalusian Spanish. Journal of the International Phonetic Association, 42,
49-63.
Perceptual assimilation of English vowels and diphthongs by Chilean Spanish speakers
Juli Cebrian (Universidad Autonoma de Barcelona, Spain)
[email protected]
The degree of similarity between existing native language (L1) categories and second/foreign language (L2) sounds
plays a crucial role in the categorization of target L2 sounds. In some cases, L1-L2 differences in inventory size result in
two or more L2 sounds mapping onto a single L1 category. Spanish and English differ greatly in vowel inventory size,
and cross-language comparisons typically involve the whole set or a subset of the English vowels and the five Spanish
monophthongs (e.g., Flege, Munro & Fox, 1994; Imai, Flege & Wayland, 2002; Iverson & Evans, 2007). However, such
an approach ignores the possible role of Spanish diphthongs in crosslinguistic perception (cf. Escudero & Williams,
2011). Further, the choice of target language variety may influence the results of cross-language perceptual mapping
(e.g., Escudero & Chládková, 2010). The goal of the current study is to evaluate the role of Spanish diphthongs in the
perceptual assimilation of English vowels to Spanish vowels. The study also explored if cross-language mappings are
influenced by the choice of target language variety. A group of 27 Chilean Spanish-speaking learners of English
performed two perceptual assimilation tasks in which listeners identified Southern British English and Canadian English
stimuli in terms of Spanish vowel categories and provided goodness-of-fit ratings. Results showed that, while a few pairs
of English vowels were mapped onto the same Spanish vowel (e.g., English /i-I/-Spanish /i/), each English diphthong
tested was consistently assimilated to a Spanish diphthong rather than to a monophthong. Despite common general
trends across the two varieties of English, some differences also emerged. Taken together, the results show that vowel
sequences like Spanish diphthongs should be included in cross-linguistic comparison studies, and that cross-language
similarity mappings may vary depending on the target language variety tested. Possible talker effects and additional
language situations are also discussed.
Escudero, P., Chládková, K. (2010). Spanish listeners' perception of American and Southern British vowels. J. Acoust.
Soc. Am. Express Letters 128(5), EL254-260.
Escudero, P., Williams, D. (2011). Perceptual assimilation of Dutch vowels by Peruvian Spanish listeners. J. Acoust.
Soc. Am. Express Letters 129(1), EL1-7.
Flege, J.E., Munro, M.J., Fox, R.A. (1994). Auditory and categorical effects on cross-language vowel perception. Journal
of the Acoustical Society of America, 95: 3623-3641.
Imai, S., Flege, J. E., Wayland, R. (2002). Perception of cross-language vowel differences: A longitudinal study of native
Spanish learners of English. Journal of the Acoustical Society of America, 111, 2364.
Iverson, P., Evans, B.G. (2007). Learning English vowels with different first-language vowel systems: Perception of
formant targets, formant movement, and duration. Journal of the Acoustical Society of America 122(5), 2842-2854.
KEYNOTE SPEAKER
Some psycholinguistic considerations on methodological issues related to speech perception tests
Márcia Cristina Zimmer (Ritter dos Reis University Center, Brazil)
This talks departs from the concept and phenomenon of categorical perception in neuroscience studies of speech
perception to review controversial methodological issues involved in speech perception tests, such as AX, Speeded AX,
ABX (and its variants AXB, XAB), 2AFC, 4IAX. Some of the tests’ main psycholinguistic constraints reported in the
literature are related to memory taxing and recency effects. However, some of the advantages of these tests are also
highlighted.
International Workshop on Cross-Language Speech Perception – University of Minho – January 30-31, 2014
-14-